Diagnostic accuracy of genetic markers for identification of the Lr46/Yr29 “slow rusting” locus in wheat (Triticum aestivum L.)

Introduction

Genetic mapping is used for the identification of the locus of a gene as well as for the determination of the distance between two genes or between a gene and a marker. Gene mapping is considered as the common breeding tool in which molecular markers are used. The principle of genetic mapping is chromosomal recombination during meiosis, which results in the segregation of loci. DNA sequences present close to the gene of interest on the same chromosome are known as linked markers. A marker without recombination to the locus of gene of interest, ideally drawn directly from the gene sequence, is described as a perfect marker [1]. Moreover, markers that could be used in multiple genetic backgrounds, ideally the marker–trait association that is valid in all germplasm, can be considered as diagnostic markers [1]. Since the 1980s, several types of molecular markers have been used in plant breeding, including random amplified polymorphic DNA, restriction fragment length polymorphism, amplified fragment length polymorphism, microsatellite or simple sequence repeats, and cleaved amplified polymorphic sequences (CAPS) [2]. Molecular markers have great potential to improve the efficiency and precision of conventional plant breeding via marker-assisted selection (MAS). This approach has been successfully practiced all over the world to supplement conventional breeding programs of plants such as wheat [3], maize [4], barley [1], and soybean [5]. Mackill and Ni (2000) and Mohler and Singrun (2004) delineated five main considerations for the use of DNA markers in MAS: reliability, quantity and quality of DNA required, technical procedure for marker assay, level of polymorphism, and cost. One of the most challenging issues is to examine the utility of molecular markers in various genetic backgrounds [6], especially breeding lines or cultivars.

Leaf rust is one of the major foliar diseases in majority of wheat-growing zones of the world. The disease is caused by obligate biotrophic fungus Puccinia triticina Erikss. & Henn. (Pt), which causes serious losses in crops every year in all wheat-growing areas of the world [7,8,9,10,11]. However, the incidence of leaf rust epidemics can be controlled through the surveillance of resistant pathogens, the development and distribution of resistant cultivars, and the judicious use of fungicides [12]. The population of Puccinia triticina consists of many physiological races with varying levels of virulence due to the airborne nature of the fungus and frequent mutation/selection. Therefore, a pathogen can rapidly develop virulence against varieties with one or more resistance genes that have been effective against the races of the pathogen found in the region [13]. It is reported that vertical resistance caused by a specific leaf rust resistance (Lr) gene does not last longer than 5–7 years [14], so new sources of resistance should be constantly searched for and introduced into wheat breeding programs.

More than 80 Lr genes have been identified and most of them were already mapped on specific wheat chromosomes using DNA markers [15]. Some of them are known as slow rust resistance genes, which are race nonspecific and confer durable adult plant resistance against multiple fungal disease in wheat [16]. The most important genes belonging to the category of “slow rusting” or APR include Lr34/Yr18/Sr57/Ltn1 [17], Lr46/Yr29/Sr58/Ltn2 [18], and Lr67/Yr46//Sr55/Ltn3 [19], which confer partial resistance to leaf rust, yellow rust, and stem rust. These genes are associated with flag leaf tip necrosis (LTN), a post-flowering morphological trait [20,21].

Singh et al. [18] first identified APR gene Lr46 on chromosome 1B in the wheat cultivar Pavon76. The Lr46 co-segregates with stripe rust resistance (Yr29) [18], stem rust resistance (Sr58) [22], powdery mildew resistance (Pm39) [23], and LTN (Ltn2) [24]. The effect of expression of Lr46 gene is smaller than of Lr34, but Lr46 gene is more effective at cooler temperatures [25]. The mechanism of slow rusting genes is still not well understood. Preliminary studies by CSIRO Plant Industry, Australia reported that the molecular mechanisms of Lr34 and Lr46 are different, and that Lr46 gene does not encode an ABC transporter protein like Lr34 [26]. So far, the sequence of the Lr46 gene has not been published, despite very intensive work in this area. Lagudah (personal communication 2020) identified several candidate genes that confer resistance to mature wheat plants. To date, several markers have been used in the research to identify the Lr46 gene in wheat: Xwmc44 [27], Xgwm259 [28], Xbarc80 [29], and csLV46G22 (E. Lagudah, unpublished data), but there are no data available of comparative efficacy markers in correlation to the distance of the marker locus from the gene on the chromosome and to field observations.

The objective of the present study was to compare four molecular markers for the identification of the Lr46 gene. The study was carried out on the wheat accessions, which were described by National Small Grain Collection (Agricultural Research Station in Aberdeen, WA, USA) as the source of slow rusting genes.

Materials and methods

Results and discussion

The presented work aimed at showing the differences between the results of the Lr46/Yr29 locus identification using four markers (Xwmc44, Xgwm259, Xbarc80, and csLV34G22) located at different distances from the gene locus. In 2003, Suenaga et al. [29] determined that the microsatellite locus Xwmc44 marker is located 5.6 cM proximal to the putative QTL for Lr46. The authors scored genotypic effects of marker loci, Xwmc44 (Lr46/Yr29) and Xgwm295 (Lr34/Yr18), on leaf rust resistance QTLs and found out that the two genes did not work additively [29]. Furthermore, Xwmc44 marker was reported as diagnostic and completely linked with Lr46 gene [30,31,32]. In our experiment, the specific product for the marker Xwmc44 with a size of 242 bp was present in 38 out of 73 tested genotypes (Table 1 and Figure 1). Another microsatellite marker, Xgwm259, was also analyzed. The locus of marker maps approximately 20 cM distal to Lr46 [26]. According to the literature, the expected product should be 105 bp [26]. Due to the large number of nonspecific products of similar size to the amplified specific product, it was difficult to evaluate by standard electrophoresis using 2% agarose gel (Figure 2). A similar problem emerged with the identification of the Xbarc80 marker. Microsatellite locus Xbarc80 maps 10–11 cM distal to Xgwm259. The expected marker product according to Giffey et al. [33] is 100 bp; however, according to the MASwheat database [34] and the marker electrophoresis attached therein, the expected product for the reference genotype is above the 100 bp standard. According to our analyses, the specific product identified in the Pavon F76 reference genotype was also slightly greater than 100 bp (Figure 3). Based on multiple repetitions of the Lr46/Yr29 locus identification using both the marker Xgwm259 and Xbarc80, we assessed the presence of the markers linked with the resistance allele. We identified a product specific for Xgwm259 marker in 52 out of 73 analyzed genotypes (Table 1 and Figure 2), while the product of Xbarc80 marker appeared in 11 genotypes (Table 1 and Figure 3). The last analyzed marker was csLV46G22, which is tightly linked to the Lr46 gene (Lagudah, pers. comm. 2020). As a result of the analyses, the marker locus was identified in 60 out of 73 analyzed genotypes (Table 1). It was reported that csLV46G22 is highly reliable and close to 100% diagnostic marker for the Lr46 gene [35,36]. This marker was used in many studies on the identification of the Lr46 gene in wheat [37–40] and triticale [41,42]. However, the primer sequences and protocols are not published by the developer, hence the marker cannot be considered for MAS, so far. Similarly, Huerta-Espino et al. [21] used two SNP markers (Viccars, L., Chandramohan, S., and Lagudah, E. unpublished data), which were located in the proximity of Lr46, with the purpose to screen a collection of bread wheat cultivars from Mexico. The other markers are located at a greater distance from the gene, and the obtained significant differences in the results indicate the unsuitability of the markers for the identification of the Lr46 gene. Our results of the identification of the Xwmc44 marker coincide 52% with the csLV46G22 marker, 75% with the Xgwm259 marker, and 32% with the Xbarc80 marker. Only 12% of the genotypes achieved the same result for all markers (Table 1). To validate and confirm the occurrence of L46 gene, Liu et al. [43] used the following molecular markers: Xwmc44, Xgwm259, and Xbarc80, jointly. Kolmer et al. [44] used the F₆ recombinant inbred lines (RILs) “Thatcher”*3/“CI13227” with csLV46G22 marker to map the 1BL chromosome region that was highly associated with resistance to multiple pathogens. The authors reported that csLV46G22 marker identified the leaf rust, stripe rust, and powdery mildew resistance at significant level [nr]. However, in the present study this marker appeared to be negative (together with other three markers) for genotype no. 31 (HI 617; PI 422283; Sujata), which was reported as a parental to carry Lr46/Yr29 gene [45,46]. According to Lan et al. [45], the Lr46/Yr29 gene was detected in the Avocet YrA × Sujata RIL population. It can lead to the hypothesis that the linkage between available markers and Lr46/Yr29 loci can be broken. What is more interesting in our another study (not published) is that we confirmed that genotype HI617 – Sujata possesses Lr67 gene, which was also reported by Lan et al. [45].

Figure 1

Electropherogram showing the presence of molecular marker Xwmc44 in tested genotypes. M, FastGene 50 bp DNA Ladder (NIPPON Genetics EUROPE GmbH); 1–73 – wheat genotypes.

Figure 2

Amplification products of PCR of wheat genotypes with Xgwm259 marker linked to Lr46 locus. M, FastGene 50 bp DNA Ladder (NIPPON Genetics EUROPE GmbH); 1–73 – wheat genotypes.

Figure 3

Electropherogram showing the presence of Xbarc80 marker linked to Lr46 gene in tested genotypes. M, FastGene 50 bp DNA Ladder (NIPPON Genetics EUROPE GmbH); 1–73 – wheat genotypes.

Considering the results of molecular analyses, it should be stated that csLV46G22 marker can be considered as the reliable positive marker for the identification of the Lr46 gene, but the analyses should be supported by the additional screening of Xwmc44 marker. The csLV46G22 CAPS marker is codominant and appeared to be useful for MAS in wheat breeding programs after publishing the primer sequences and protocols. It should be mentioned that scoring of LTN, a morphological trait (Ltn2), which is closely linked to Lr46/Yr29 loci [23], can also be considered for selection of resistant wheat genotypes.

In summary, it can be said that the extended durability of the Lr46 gene seems to be suitable for attempting to clone this gene or genes based on the previous studies of wheat multipathogen APR genes Lr34/Yr18/Sr57/Pm38 and Lr67/Yr46/Sr55/Pm46. Cobo et al. mapped the 1BL chromosome region which overlaps with Lr46/Yr29 loci [35]. These authors identified 13 genes in the candidate region that are annotated with functions associated with disease resistance. The latest maps for Lr46/Yr29/ from Pavon 76 place this locus between TraesCS1B01G453900 and csLV46G22 (Lagudah, unpublished data, 2018), a region very similar to the candidate region of the 332 kb gene for QYr.ucw-1BL identified by Cobo et al. [35]. Tomkowiak et al. [47] reported the differences in the expression of microRNAs (miR5085 and miR164) associated with the Lr46 gene and proved that miR164 can be involved in leaf rust resistance mechanisms. Detailed identification of Lr46/Yr29 region sequence will be a key issue for characterizing the allelic variation presence in breeding collections of wheat and become a significant support for MAS programs.